The application of faecal egg count results and statistical inference for clinical decision making in foals.

The research article presents a study on the variability of faecal egg counts (FEC) in foals and its impact on veterinary decision-making regarding treatment of parasitic infections. The main finding suggests that a single estimate of FEC may not be reliable due to considerable variation; therefore employing confidence intervals from multiple counts is recommended for more accurate evaluation and treatment decisions.

Study Methodology

• The research was carried out on 53 foals. Each foal provided a single faecal sample daily for three consecutive days.
• From each daily sample, three separate portions were analyzed for faecal egg count (FEC), resulting in a total of nine egg counts per foal.
• The fecal samples were evaluated for presence and count of Parascaris spp. and strongyle eggs, common intestinal parasites in horses.

Analysis and Decision Making

• The decision to administer anthelmintic treatment (medication to expel parasitic worms) was assessed based on different statistical benchmarks: a single count, the mean count of three samples, the mean count of nine samples and the upper 5% confidence limit of the gamma confidence interval (`CI`) of the mean count from either three or nine samples.
• The gamma confidence interval was used as the best estimate for the mean egg count distribution. Comparisons were made using these different benchmarks to evaluate decision-making

Study Findings

• The results showed that point estimates (i.e., a single count or the mean of three counts) were unreliable for estimating the distribution of egg counts in faecal samples. This is due to the inhomogeneous nature of egg distribution, resulting in significant variability between samples, portions of the same sample, and different days.
• The study found poor agreement in treatment decisions when using a single count compared with using the upper confidence limit of the mean count from multiple counts.
• Even in cases where the single count showed zero eggs per gram, in 54% and 47% of cases, respectively, subsequent counts revealed presence of Parascaris and strongyle eggs.

Conclusion

• Given the significant variability in faecal egg counts in foals, use of the negative binomial distribution confidence interval for estimating the mean count is recommended. This approach accounts for variability and can provide a more reliable base for interpretation of FEC data and subsequent treatment decision-making in veterinary practice.